Research report
Which patients receive antidepressants? A `real world' telephone study

https://doi.org/10.1016/S0165-0327(97)00193-6Get rights and content

Abstract

Introduction: The use of antidepressants has been questioned with respect to both undertreatment and inadequate prescription. The present investigation was therefore launched to assess the psychopathology profiles of antidepressant users. Methods: A representative sample was constituted on the basis of usual antidepressant consumption, and ICD 10 compatible diagnoses were obtained after telephone administration of a structured psychiatric interview. Results: The most often used drugs were fluoxetine, followed by tricyclic antidepressants. Coprescription existed in slightly less than two thirds of antidepressant users. ICD 10 diagnoses were compared to currently available prescription guidelines. Fluoxetine prescription, as compared to other drugs, was found to be significantly more compliant with these guidelines; conversely, in 22% of antidepressant users, no complete ICD 10 diagnosis could be documented. These results are discussed in the light of report accuracy and anecdotal or `heterodox' indications of antidepressants. Conclusion: Altogether, the present study confirms previous doubts regarding appropriate use of antidepressants and stresses the need for more explicit and comprehensive clinical guidelines. It does not substantiate, however, any evidence for a `recreational' use of these products.

Introduction

Antidepressants have been used for over three decades and their clinical efficacy and indications are largely consensual. There remain a number of troublesome questions, however, regarding their actual use outside the specific setting of clinical trials. The undertreatment of clinically diagnosed depressed patients has been repeatedly reported (Isacsson et al., 1992, Wells et al., 1994). Recent concerns also appeared after newer antidepressants such as specific serotonin reuptake inhibitors (SSRI) were marketed: firstly, a number of pharmacoeconomic papers have dealt with increasing health costs due to the higher price of the more recent drugs, although it is increasingly recognized that this situation is overall balanced by their better acceptability, which might lower the overall health costs associated with depression (Montgomery and Kasper, 1995, Simon et al., 1995a). Secondly, there has been a wide debate in the lay as well as in the academic press over personality changes and extratherapeutic or recreational use of these drugs (Kramer, 1993, Barondes, 1994, Medawar, 1994, Freeman, 1994, Kaplan, 1994, Walder, 1994, Manolis, 1995).

The classical, `gold standard', controlled clinical trials are poorly, if at all, able to deal with such problems, which are better addressed by epidemiological, pharmacoeconomic or `real world' studies, respectively. The advantages of the latter are obvious in terms of assessment of actual patterns of prescription (Simon et al., 1995b).

The present work was undertaken to provide real world data regarding the use of antidepressant medications, with special emphasis on current diagnosis, treatment history, and overall appropriateness of the prescription with respect to current guidelines, in a representative sample of the French population. This sample, which will be described in more detail in the methods section, was selected on the basis of antidepressant use, whatever the diagnosis, which overcomes the bias induced by patient or physician report. The diagnosis was established retrospectively by the means of a structured telephone interview.

Section snippets

The SOFRES database.

SOFRES, a poll company, routinely scrutinizes a sample group of the French population, comprising over 40 000 individuals in 20 000 households. This sample group has been shown to be representative of the French population as a whole with respect to age, social status, size of the household, and living area. It is renewed by a quarter each year.

Step 1. A questionnaire aimed at identifying antidepressant consumption at the time of the study was sent to all households, out of which 15 800

Overall prescription pattern

Eight hundred and ninety-six individuals (2.75%) reported to be on antidepressant treatment at the time of the study. Extrapolation of this figure to the whole population suggests that 1 094 300 persons received antidepressive drugs (ADD) at that time in France.

Antidepressant users were more likely to live in the south east of France, and less likely in the north and east as well as in a rural area (P<0.05). The sex ratio (M/F) was 27/73 as opposed to 48.1/51.9 in non users (P<0.05). People

Discussion

There is little doubt that actual drug use seldom strictly complies with ad hoc clinical guidelines. This situation stresses the need for investigation tools able to account for such discrepancies. Recently `real world studies' have turned out to be of special interest in this special context and their advantages have been underscored (Simon et al., 1995b). The present study attempted such an investigation, with the aim of assessing to what extent antidepressants prescription conforms to

Conclusion

Overall, the present study confirms previous doubts regarding the appropriate use of antidepressants. The high rate of concomitant prescription of benzodiazepines, and the fact that 22% of users appear not to pertain to any psychiatric diagnosis appear especially troublesome. In this context, the prescription of fluoxetine appears to be significantly more consistent with acknowledged guidelines than that of other antidepressants. In spite of these findings, it must also be stressed that the

References (19)

There are more references available in the full text version of this article.

Cited by (21)

View all citing articles on Scopus
View full text